For the first time in 15 years, the marketing technology landscape has effectively stopped growing. The State of Martech 2026, released on 5 May by Scott Brinker of chiefmartec and Frans Riemersma of MartechTribe, counts 15,505 products — up just 121 on last year, or 0.79%. After a run that took the landscape from 150 tools in 2011 to more than 15,000, the annual figure the industry has watched as a scoreboard for a decade and a half barely moved.

That near-zero figure grabs the attention, but it is not the story. Beneath it, the market is turning over: 1,488 products were added over the year and 1,367 were removed, the first time in the post-pandemic era that new arrivals and departures have very nearly cancelled each other out. Brinker and Riemersma argue this is not the market freezing but the market metabolising — and that the practical consequence for marketers is a shift from buying tools to swapping them.

What the flat number is hiding

The State of Martech 2026 frames the plateau as a surface reading of a market in motion. New entrants fell about 40% year on year, down from 2,489 the previous year, while removals rose roughly 13%. The result is near-parity between tools born and tools retired — a pattern the report reads as renewal rather than stagnation, with weaker or ageing products cleared out as new, mostly AI-native ones take their place.

That churn matters more to a marketing team than the top-line count. A near-flat landscape that is quietly replacing a large share of itself each year is one in which the tool a team bought two years ago is likelier than before to be acquired, merged or shut down. The report’s authors resist the tidy “peak martech” conclusion the flat number invites; Brinker has said the field may finally have hit a ceiling, but the report’s own position is that the landscape is renewing, not topping out.

Why the report calls it a ‘Darwin phase’

Brinker and Riemersma describe martech as entering a “Darwin phase”: an environment in which the cost of building keeps falling, AI keeps opening new niches, and the minimum scale needed to sustain a martech business keeps shrinking, so new kinds of tool appear and unfit ones die off faster. The era of accumulating tools, they argue, is giving way to an era of replacing them.

Underneath that is a structural change in where value sits. The report contends that SaaS platforms are no longer the main source of differentiation and are becoming infrastructure — systems of record, workflow engines and integration layers that provide stability while the competitive action moves up to an orchestration and agent layer built on top. The rapid spread of the Model Context Protocol is offered as evidence: the report points to more than 29,000 MCP servers listed across registries in roughly 18 months, a sign of how quickly the connective tissue between tools and AI is being laid down.

What it means for a marketing team’s stack

Brinker’s sharpest warning is that AI is exposing the stack. As AI moves from generating content to orchestrating work across tools, the weaknesses teams have long tolerated — poor data quality, weak governance, brittle integrations — stop being back-office irritations and become visible risks to the customer experience. An agent acting on bad data does so quickly and at scale.

That reframes the stack as the thing to scrutinise, and gives marketers cover to do what many have wanted to for years: stop adding and start pruning. The signal from the data is to judge a tool less on its feature list than on whether it will survive the shakeout and how cleanly it integrates with everything else. Composability — the ability to swap one component without rebuilding the whole system — is, on the report’s reading, what keeps the steadier vendors steady.

It also moves the marketer’s own job up a level. The skill the report implies is less about operating any single tool than about orchestration: defining how tools, data and agents work together, and engineering the context they draw on. That is a more demanding remit than managing a stack of point solutions, and a more durable one.

The count that defined a martech era may never again be the number everyone watches. The open question the report leaves is which of today’s incumbents adapt fast enough to remain infrastructure and which become this phase’s casualties — a question every team answers, in practice, each time it renews a contract.